Futoshi Futami
Futoshi Futami develops mathematical methods for probabilistic machine learning and uncertainty estimation. His research includes Bayesian inference and the calibration of predictions. Futami, Issei Sato and Masashi Sugiyama coauthored research on variational inference based on robust divergences. The paper appeared at AISTATS 2018. Futami and Masahiro Fujisawa studied time-independent information-theoretic generalization bounds for stochastic gradient Langevin dynamics. The work appeared at NeurIPS 2023.
Timeline
-
2018
Robust variational inference
Futami, Issei Sato and Masashi Sugiyama coauthored research on variational inference based on robust divergences. The paper appeared at AISTATS 2018.
-
2023
Generalization in stochastic learning
Futami and Masahiro Fujisawa studied time-independent information-theoretic generalization bounds for stochastic gradient Langevin dynamics. The work appeared at NeurIPS 2023.
-
2024
Understanding confidence calibration
Futami and Fujisawa coauthored an information-theoretic analysis of expected calibration error. The work examines a measure used to compare predicted probabilities with observed outcomes.
-
As of October 2026
Uncertainty Quantification Team
Futami directs RIKEN’s Uncertainty Quantification Team.
Sources and credits
Photo: futoshi-futami.github.io; creator not stated. License: Not stated by source. Source: futoshi-futami.github.io. Cropped to a square and resized. Ask us to remove or replace this photo.